Script analytics to generate quality score and report

Abstract

Embodiments provide for evaluation of scripts. A script for producing media content is received, and a plurality of tags related to content of the script is determined. A quality score is generated for the script by processing the plurality of tags using a first model, and one or more modifications for the script are generated based on the quality score and the plurality of tags.

What is claimed is: 
     
       1. A computer-implemented method, comprising:
 receiving, by a processor:
 a script for producing media content, 
 
 a plurality of tags related to content of the script, wherein the plurality of tags are generated using a first machine learning model trained to generate tags based on script elements, and wherein the script elements comprise plot devices, character types, or dialogue counts or distributions, and 
 a quality score for the script generated by a second machine learning model configured to process the plurality of tags, wherein a selection of the second machine learning model is based at least in part on a genre of the script; 
 outputting, by the processor and via a graphical user interface, one or more proposed modifications for the script based on the quality score and the plurality of tags; 
 automatically applying the one or more proposed modifications to the script; and 
 automatically refining the second machine learning model based on feedback relating to the quality of the script. 
 
     
     
       2. The method of  claim 1 , wherein generating the plurality of tags further comprises processing the script using one or more trained machine learning models in order to identify occurrence of one or more predefined concepts in the script. 
     
     
       3. The method of  claim 1 , further comprising receiving one or more manually-curated tags from a user, wherein the quality score is further based on the one or more manually-curated tags. 
     
     
       4. The method of  claim 1 , wherein the plurality of tags indicate at least one of:
 (i) presence of one or more plot devices in the script; 
 (ii) presence of one or more character archetypes in the script; 
 (iii) gender of one or more characters in the script; 
 (iv) a diversity of characters in the script; 
 (v) a number of lines associated with each character in the script; 
 (vi) a length of the script; 
 (vii) a complexity of dialogue in the script; or 
 (viii) a length of one or more segments of the script. 
 
     
     
       5. The method of  claim 1 , wherein the one or more proposed modifications are generated responsive to comparing the quality score to a threshold quality score, and wherein generating the one or more proposed modifications comprises identifying one or more actions that, if performed, would increase the quality score of the script. 
     
     
       6. The method of  claim 1 , wherein the one or more proposed modifications comprise a suggestion of a particular delivery mode for the produced media content in order to improve probability that the produced media content will be successful. 
     
     
       7. The method of  claim 1 , further comprising:
 training the second machine learning model to generate quality scores, wherein the training uses a set of example scripts and associated quality scores. 
 
     
     
       8. The method of  claim 1 , wherein the quality score is based at least in part on a number of plot devices, a timing of plot devices, types of plot devices, or combinations thereof. 
     
     
       9. The method of  claim 8 , wherein the types of plot devices include character reveals, audience reveals, reversals, or cliffhangers. 
     
     
       10. The method of  claim 8 , wherein the quality score is based at least in part on a weight associated with a segment of the script in which a plot device of the plot devices occurs. 
     
     
       11. The method of  claim 1 , wherein generating the quality score includes applying natural language processing to determine complexity of dialogue in the script. 
     
     
       12. A non-transitory computer-readable medium containing computer program code that, when executed by operation of one or more computer processors, performs an operation comprising:
 receiving a script for producing media content; 
 automatically generating a plurality of tags related to content of the script, wherein the plurality of tags are generated using a first machine learning model trained to generate tags based on script elements, and wherein the script elements comprise plot devices, character types, or dialogue counts or distributions; 
 selecting a second machine learning model, wherein the second machine learning model is selected from a plurality of models based at least in part on a genre of the script; 
 generating a quality score for the script by processing the plurality of tags using the selected second machine learning model; 
 generating, based on the quality score and the plurality of tags, one or more modifications for the script; 
 outputting, via a graphical user interface, the quality score and the one or more modifications for the script; 
 automatically applying the one or more modifications to the script; and 
 automatically refining the second machine learning model based on;
 a determination that the quality score is above or below a threshold, 
 a popularity of the media content based on the modified script, 
 a delivery mode of the media content, and 
 adjusting one or more predefined thresholds used by the second machine learning model to generate the quality score. 
 
 
     
     
       13. The computer-readable medium of  claim 12 , wherein generating the plurality of tags further comprises processing the script using one or more trained machine learning models in order to identify occurrence of one or more predefined concepts in the script. 
     
     
       14. The computer-readable medium of  claim 12 , wherein the operation further comprises receiving one or more manually-curated tags from a user, and wherein the quality score is further based on the one or more manually-curated tags. 
     
     
       15. The computer-readable medium of  claim 12 , wherein the plurality of tags indicate at least one of:
 (i) presence of one or more plot devices in the script; 
 (ii) presence of one or more character archetypes in the script; 
 (iii) gender of one or more characters in the script; 
 (iv) a diversity of characters in the script; 
 (v) a number of lines associated with each character in the script; 
 (vi) a length of the script; 
 (vii) a complexity of dialogue in the script; or 
 (viii) a length of one or more segments of the script. 
 
     
     
       16. The computer-readable medium of  claim 12 , wherein the one or more modifications are generated responsive to comparing the quality score to a threshold quality score, and wherein generating the one or more modifications comprises identifying one or more actions that, if performed, would increase the quality score of the script. 
     
     
       17. A system, comprising:
 one or more computer processors; and 
 a memory containing a program which when executed by the one or more computer processors performs an operation, the operation comprising:
 receiving a script for producing media content; 
 automatically generating a plurality of tags related to content of the script, 
 
 wherein the plurality of tags are generated using a first machine learning model trained to generate tags based on script elements, and wherein the script elements comprise plot devices, character types, or dialogue counts or distributions;
 selecting a second machine learning model, wherein the second machine learning model is selected from a plurality of models based at least in part on a genre of the script; 
 generating a quality score for the script by processing the plurality of tags using the selected second machine learning model; 
 generating, based on the quality score and the plurality of tags, one or more modifications for the script; 
 outputting, via a graphical user interface, the quality score and the one or more modifications for the script; 
 
 automatically applying the one or more modifications to the script; and 
 automatically refining the second machine learning model based on:
 a determination that the quality score is above or below a threshold, 
 a popularity of the media content based on the modified script, 
 a delivery mode of the media content, and 
 adjusting one or more predefined thresholds used by the second machine learning model to generate the quality score. 
 
 
     
     
       18. The system of  claim 17 , wherein generating the plurality of tags further comprises processing the script using one or more trained machine learning models in order to identify occurrence of one or more predefined concepts in the script. 
     
     
       19. The system of  claim 17 , wherein the plurality of tags indicate at least one of:
 (i) presence of one or more plot devices in the script; 
 (ii) presence of one or more character archetypes in the script; 
 (iii) gender of one or more characters in the script; 
 (iv) a diversity of characters in the script; 
 (v) a number of lines associated with each gender in the script; 
 (vi) a length of the script; 
 (vii) a complexity of dialogue in the script; or 
 (viii) a length of one or more segments of the script. 
 
     
     
       20. The system of  claim 17 , wherein the one or more modifications are generated responsive to comparing the quality score to a threshold quality score, and wherein generating the one or more modifications comprises identifying one or more actions that, if performed, would increase the quality score of the script.